Correlation between Coronavirus (COVID-19) Vaccinations and Excess Mortality in the OWID Dataset

Get and Process the Data

We get the data from OWID and interpolate/upsample the missing values weekly.

Filter Out Countries With Insufficient Data

If there is no single row with both Vaccinations and Excess Mortality data we have nothing to work with. So we filter such countries out completely.

Interpolate The Data By Backfilling

Some countries have the data reported monthly while others have missing data points. We resample everything to the weekly grane and interpolate via backfilling. This way the vaccination time series before the start of reporting are backfilled with a static value (first value in time serie), mostly close to zero. This makes the year 2020 kind of a synthetic control in regard to the vaccination treatment.

Overall Correlation between Vaccinations and Excess Mortality

At the moment of writing the correlation is close to zero (-0.04). Let's de-average the overal value by calculating the correlation per country.

De-averaging Of The Correlation By Country

We chart correlation between excess_mortality and total_vaccinations_per_hundred (and other vaccination metrics) per country to make sure the data is sane.

The Choropleth Charts give an idea how the correlation is distibuted over the globe. May be there is some role the seasonality plays in the resulting in-country corellation.

The Histograms show that the in-country correlation is kinda evenly distributed over the whole [-1:1] interval and there are no visible ouliers.

Interpolated Data For Visual Control

Obligatory XKCD parabole

Obligatory XKCD parabole

Permanent link to this comic: https://xkcd.com/925/

Vaccine is present/absent

Correlation between excess_mortality and total_vaccinations_per_hundred in countries where a vaccine is present/absent

For each vaccine we split all countries into two groups as per vaccine presence and calculate the correlation within those two group. Obviously there are countries with multiple vaccine present. And some vaccine may only come together in all countries. Yet the results are interesting and non-trivial.

Copyright and Getting the Data

Copyright 2021 Abbrivia GmbH https://www.abbrivia.com CC-BY (By Attribution) 4.0 https://creativecommons.org/licenses/by/4.0/legalcode Reuse our work freely!

All visualizations, and code produced in this notebook are completely open access under the Creative Commons BY license. You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.

The data produced by third parties and made available by "Our World in Data" is subject to the license terms from the original third-party authors. Check the license of any third-party data before use and redistribution on 'https://ourworldindata.org/coronavirus' site (see below).

See the defintions and further discussion on the used dataset at the "Our World in Data" site https://ourworldindata.org/covid-vaccinations

The data is taken specifically from https://covid.ourworldindata.org/data/owid-covid-data.xlsx file

Hannah Ritchie, Edouard Mathieu, Lucas Rodés-Guirao, Cameron Appel, Charlie Giattino, Esteban Ortiz-Ospina, Joe Hasell, Bobbie Macdonald, Diana Beltekian and Max Roser (2020) - "Coronavirus Pandemic (COVID-19)". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/coronavirus' [Online Resource]

We use Excel file because it contains the data format information in itself. If you want to run this more often consider manually downloading the data and sourcing it locally as shown in the next line (commented out).